CN112581188A - Construction method, prediction method and model of engineering project bid quotation prediction model - Google Patents

Abstract

The invention discloses a construction method, a prediction method and a model of an engineering project bid quotation prediction model. The construction method of the prediction model comprises the following steps: and extracting the characteristics of the original data with a plurality of characteristic variables by adopting a principal component analysis or hierarchical clustering method, and constructing a multiple linear regression model by using the extracted data corresponding to a small number of characteristic variables. And determining an optimal prediction model by using the R-squared value, the P value and the average absolute error MAE, and carrying out hypothesis test on the model. And finally, predicting the bid quotation in the actual engineering by using the model. The prediction model constructed by the method has the advantages of less data acquisition amount, high prediction speed, time and labor saving, higher accuracy and good reference value, and the data amount is usually more than 30.

Description

Construction method, prediction method and model of engineering project bid quotation prediction model

Technical Field

The invention belongs to the technical field of project bid quotation prediction, and particularly relates to a construction method, a prediction method and a model of a project bid quotation prediction model.

Background

Bid quotes are a complex and comprehensive technical and economic task. Whether the quote is accurate or not is critical in correctly calculating the cost. Due to the characteristics of singleness, type diversity, price advancement, synchronization of production and sales, etc., of building products, bidding offers are focused on the future, and prediction is critical.

The existing project price forecast is estimated by the way of project cost: and B, pre-estimating the bid price according to the drawing, the engineering quantity and the like provided by the A party. The existing mode is adopted, and time and labor are wasted.

Disclosure of Invention

The invention provides a construction method, a prediction method and a model of a prediction model of an engineering project bid price, aiming at solving the problems of large workload and large consumption of manpower caused by manually providing a data set quota and budgeting according to a Party A in the existing engineering project bid price.

The invention is realized by the following technical scheme:

a construction method of a prediction model of engineering project bid quotation comprises the following steps:

feature extraction is carried out on the data by utilizing Principal Component Analysis (PCA) or hierarchical clustering,

constructing an initial multiple linear regression model,

screening the initial model by utilizing the R-squared value, the P value and the average absolute error MAE and establishing an optimal prediction model;

and performing hypothesis test on the optimal model.

A road and bridge project bid quotation prediction model is constructed by adopting the method, and the prediction model is as follows:

Avgoffer＝investment+investment:bridgeProp，

the Avgoffer is average quoted price, the inventment is limit price, the bridge Prop is bridge proportion, and the inventment is the interaction between the inventment and the bridge Prop.

A road and bridge project bid quotation prediction method comprises the following steps:

acquiring price limit inventment and bridge proportion bridge Prop data;

and substituting the price limit inventment and the bridge proportion bridgeProp data into the prediction model.

Compared with the prior art, the invention at least has the following advantages and beneficial effects:

1. the prediction model constructed by the method has the advantages of less acquired data, high prediction speed, time and labor saving and higher accuracy, and the data volume is usually more than 30.

2. The road and bridge project bid price forecasting model only needs to collect two data of price limit and bridge ratio, and has the advantages of small data collection amount, high forecasting speed and high forecasting accuracy.

Drawings

In order to more clearly illustrate the embodiments of the present invention or the technical solutions in the prior art, the drawings used in the description of the embodiments or the prior art will be briefly described below, it is obvious that the drawings in the following description are only some embodiments of the present invention, and for those skilled in the art, other drawings can be obtained according to the drawings without creative efforts.

FIG. 1 is a flow chart of a predictive model construction method of the present invention.

FIG. 2 is a table of data involved in a road bridge entry.

Fig. 3 is a diagram showing the result of feature extraction of data by the principal component analysis method.

Fig. 4 is a diagram showing the result of processing data by hierarchical clustering.

Fig. 5 is a graph showing the results of a linear correlation test performed on the optimal prediction model.

Fig. 6 is a diagram showing the result of a normal distribution test performed on the optimal prediction model.

Fig. 7 is a diagram showing the results of the homodyne test performed on the optimal prediction model.

Detailed Description

The invention is further described with reference to the following figures and specific embodiments. It should be noted that the description of the embodiments is provided to help understanding of the present invention, but the present invention is not limited thereto. Specific structural and functional details disclosed herein are merely illustrative of example embodiments of the invention. This invention may, however, be embodied in many alternate forms and should not be construed as limited to the embodiments set forth herein.

It will be understood that, although the terms first, second, etc. may be used herein to describe various elements, these elements should not be limited by these terms. These terms are only used to distinguish one element from another. For example, a first element could be termed a second element, and, similarly, a second element could be termed a first element, without departing from the scope of example embodiments of the present invention.

It should be understood that, for the term "and/or" as may appear herein, it is merely an associative relationship that describes an associated object, meaning that three relationships may exist, e.g., a and/or B may mean: a exists alone, B exists alone, and A and B exist at the same time; for the term "/and" as may appear herein, which describes another associative object relationship, it means that two relationships may exist, e.g., a/and B, may mean: a exists independently, and A and B exist independently; in addition, for the character "/" that may appear herein, it generally means that the former and latter associated objects are in an "or" relationship.

It will be understood that when an element is referred to herein as being "connected," "connected," or "coupled" to another element, it can be directly connected or coupled to the other element or intervening elements may be present. Conversely, if a unit is referred to herein as being "directly connected" or "directly coupled" to another unit, it is intended that no intervening units are present. In addition, other words used to describe the relationship between elements should be interpreted in a similar manner (e.g., "between … …" versus "directly between … …", "adjacent" versus "directly adjacent", etc.).

It is to be understood that the terminology used herein is for the purpose of describing particular embodiments only and is not intended to be limiting of example embodiments of the invention. As used herein, the singular forms "a", "an" and "the" are intended to include the plural forms as well, unless the context clearly indicates otherwise. It will be further understood that the terms "comprises," "comprising," "includes" and/or "including," when used herein, specify the presence of stated features, integers, steps, operations, elements, and/or components, but do not preclude the presence or addition of one or more other features, numbers, steps, operations, elements, components, and/or groups thereof.

It should also be noted that, in some alternative designs, the functions/acts noted may occur out of the order noted in the figures. For example, two figures shown in succession may, in fact, be executed substantially concurrently, or the figures may sometimes be executed in the reverse order, depending upon the functionality/acts involved.

It should be understood that specific details are provided in the following description to facilitate a thorough understanding of example embodiments. However, it will be understood by those of ordinary skill in the art that the example embodiments may be practiced without these specific details. For example, systems may be shown in block diagrams in order not to obscure the examples in unnecessary detail. In other instances, well-known processes, structures and techniques may be shown without unnecessary detail in order to avoid obscuring example embodiments.

Example 1

The average bid is the average bid price for the same project. There are many factors that affect the average price of a project, but not all characteristic variables will affect it. Therefore, a construction method of the engineering project bid quotation prediction model needs to extract the characteristics of the original data. Specifically, as shown in fig. 1, the method includes the following steps:

the raw data is subjected to feature extraction, which can be based on various methods, such as a principal component analysis method or a hierarchical clustering method.

And then, an initial prediction model is constructed according to data corresponding to the extracted variables, at this time, not all the extracted characteristic variables can be adopted, and further verification is needed to be carried out on the extracted characteristic variables so as to improve the accuracy of the prediction model.

And screening the initial model by using parameters such as the R-squared value, the P value, the average absolute error MAE and the like, and establishing an optimal prediction model.

The R-squared value refers to how much of the dependent variable is interpreted by the independent variable.

The P value is an index of whether a zero hypothesis is rejected in hypothesis test, and usually the P value of a certain characteristic variable is greater than 0.05, which indicates that the characteristic variable is not significant and does not have a good explanation on the corresponding average report value, and the characteristic variable can be removed.

The mean absolute error MAE is the average of absolute errors between the predicted value and the observed value, and the smaller the value, the closer the test value and the true value are represented, i.e. the better the model is. And comprehensively considering the R-squared value, the P value and the average absolute error MAE to determine an optimal prediction model.

And performing hypothesis test on the optimal model.

The prediction model constructed by the method has high prediction accuracy.

Example 2

The present embodiment describes the implementation of the method specifically by taking the road and bridge project as an example.

The feature extraction about the road and bridge project comprises a segment length, an altitude, a construction period, a long tunnel bigTunnel, a middle tunnel midTunnel, a short tunnel minTunnel, a tunnel length lofTunnel, a tunnel proportion tunnelProp, a big bridge bigBridge, a middle bridge midBridge, a small bridge minBridge, a bridge length lofBridge, a bridge proportion brigeProp, a lane Nofroad, a price limit inventment and an average price offer Avgoffer. The data is shown in fig. 2. The bidding quotations of bidders are generally influenced by factors such as the length of the project, the construction period, the bridge proportion and the like. That is, based on the first 15 feature variables in the above features, the last feature variable 'average offer Avgoffer' is predicted. However, the first 15 characteristic variables do not all affect the bid price, and it is necessary to screen the characteristic variables and build a prediction model to predict the bid price. Specifically, the method comprises the following steps:

and extracting characteristic variables, wherein the extracted characteristic variables comprise the bridge length lofBridge, the bridge proportion bridge Prop, the lane Nofroad, the price limit invent and the average quotation Avgoffer. The step can adopt a principal component analysis method or a hierarchical clustering method to carry out feature extraction.

The principal component analysis method performs dimensionality reduction on the characteristic variables, and the result is shown in a box A in fig. 3, the bridge length lofBridge, the bridge proportion bridgeProp, the lane Nofroad, the price limit inventment and the average quoted Avgoffer are classified into one category, wherein the coordinates of each point are detailed in the following table 1. The principal component analysis selects a small number of dimensions from a plurality of dimensions as a representation, and can restore and represent original data by using the data of the small number of dimensions.

TABLE 1

Specifically, the principal component analysis comprises the following steps:

standardizing the original data, and converting the original data into data with an average value of 0 and a standard deviation of 1;

calculating a covariance matrix of the normalized data;

calculating an eigenvalue and an eigenvector of the covariance matrix;

sorting the eigenvalues in a descending order, and then sorting the corresponding eigenvectors in a descending order;

and (5) extracting the characteristic vector corresponding to the characteristic value with the maximum first 2 values, drawing a biplot graph, and finding out the characteristic variable which is relatively close to the target variable in the graph.

And processing the data by adopting a hierarchical clustering method to obtain a result shown in the figure 4, wherein the result is consistent with the result obtained by adopting a principal component analysis method, and the details are shown in a frame B in the figure 4. Specifically, the hierarchical clustering method comprises the following steps:

the data is preprocessed, typically by converting the raw data into data with a mean of 0 and a standard deviation of 1.

And measuring the distance between a pair of data or data between two categories in the processed data, wherein the distance is an Euclidean distance or a relative distance, and the relative distance is selected in the embodiment.

The two groups of data or two types of data are aggregated according to different distance categories, wherein the distance categories generally comprise maximum distance, minimum distance, average distance and centroid distance, and the maximum distance is selected in the embodiment.

The number of categories can be obtained by traversing the dendrogram with a horizontal line. Feature variables of the same category can be classified into one class.

According to the extracted feature extraction, an initial prediction model is constructed, and multiple linear regression is carried out, wherein the model is as follows: avgoffer ═ inventment + lofBridge + bridgeProp + Nofroad;

and (3) verifying each characteristic variable in the prediction model by using a P value, wherein the P value of investment is 0.000, the P value of lofBridge is 0.962, the P value of bridgeProp is 0.049, the P value of Nofroad is 0.624, and the characteristic variables with the P values larger than 0.05 are removed, namely lofBridge and Nofroad are removed. The eliminated feature variables only remain inventment and bridgeProp, and the following 5 prediction models can be formed by considering the interaction between the feature variables.

The 5 prediction models were:

1.Avgoffer＝investment+bridgeProp，

2.Avgoffer＝investment+bridgeProp+investment:bridgeProp，

3.Avgoffer＝investment+investment:bridgeProp，

4.Avgoffer＝bridgeProp+investment:bridgeProp，

5.Avgoffer＝investment:bridgeProp，

wherein, inventment: bridgeProp represents the interaction of inventment and bridgeProp.

The R-squared value can be seen first, and the closer the R-squared value is to 1, the better the model fits. And calculating the R-squared values of the models, wherein the R-squared values of the models from 1 to 3 are all 0.996, the R-squared value of the model No. 4 is 0.789, and the R-squared value of the model No. 5 is 0.719. Models No. 4 and 5 may be excluded.

Then, the errors between the predicted values and the actual values of the 5 prediction models are tested. Specifically, the mean absolute error MAE of the predicted value and the true value in the test set may be used. And selecting a model with a smaller mean absolute error MAE value as a candidate prediction model. The smaller the MAE value, the closer the predicted value and the true value are represented, and the higher the prediction accuracy is.

The average absolute error MAE of the 5 prediction models is tested to be 1058.70, 776.20, 944.10, 15756.21 and 21732.17 respectively. Prediction models No. 2 and No. 3 are taken as candidate prediction models.

And continuously observing the P value of the candidate prediction model, wherein the P value of the bridgeProp in the No. 2 prediction model is 0.958, and the P value of the innovative prediction model, namely the bridgeProp is 0.087, does not meet the requirement that the P value is less than 0.05, so the prediction model is rejected. In model No. 3, the P value of inventment is 0.000, and the P values of inventment and bridgeProp are 0.001, which both meet the requirement that the P value is less than 0.05.

Model No. 3 was therefore selected as the optimal prediction model, namely: avgoffer ═ inventstment + inventstment: bridgeProp.

After the model is built, the model is verified,

1. performing a linear correlation test: that is, the Avgoffer, inventment and bridgeProp show a linear relationship, and as shown in FIG. 5, the SAS software is used for drawing to verify that the requirements are met.

2. Carrying out normal distribution test: the model residuals (the difference between the predicted y-value and the true y-value) are normally distributed, as tested by the Q-Q plot. If the residuals are on a straight line in the Q-Q plot, it indicates that the residuals satisfy the normal distribution. Q-Qplot of this model is shown in FIG. 6, verifying satisfaction.

3. Checking the homological difference: a scatter plot between the residual and the predicted values is made. If the dots in the graph, shown in fig. 7, are evenly distributed across the red line without exhibiting a distinct shape, it is indicated that the residual satisfies homoscedasticity. As shown in fig. 7, the model verifies satisfaction.

And (4) performing hypothesis test, wherein the model basically meets the hypothesis condition of the multiple linear regression model.

After the model is built, when an actual project is used, only data corresponding to two characteristic variables of price limit inventment and bridge proportion bridge Prop need to be collected. Inputting 3 groups of data (each group comprises two data) into the optimal prediction model, and outputting the predicted bid quotation.

Take the following 3 data sets as an example:

mark section	investment	bridgeProp
				1	77550	5.4
2	146178	100
			3	67895	15.2

The comparison between the predicted result and the actual result is shown in the table below.

Real average quotation (ten thousand yuan) after marking segment name	Average price quote for model prediction (ten thousand yuan)	Absolute difference (Wanyuan)
			1 70310.90754	70684.93638	374.02884
2 141470.1467	139365.9025	2104.2442
			3 61077.30723	62435.27912	1357.97189

Section name	True reduction of average quotes	Prediction of average quotes Amplitude reduction
			1	9.33％	8.85％
2	3.22％	4.66％
			3	10.04％	8.04％

And (4) conclusion: the predicted result is approximately close to the real result, and has a certain reference value.

The embodiments described above are merely illustrative, and may or may not be physically separate, if referring to units illustrated as separate components; if reference is made to a component displayed as a unit, it may or may not be a physical unit, and may be located in one place or distributed over a plurality of network units. Some or all of the units can be selected according to actual needs to achieve the purpose of the solution of the embodiment. One of ordinary skill in the art can understand and implement it without inventive effort.

The above examples are only intended to illustrate the technical solution of the present invention, but not to limit it; although the present invention has been described in detail with reference to the foregoing embodiments, it will be understood by those of ordinary skill in the art that: modifications may be made to the embodiments described above, or equivalents may be substituted for some of the features described. And such modifications or substitutions do not depart from the spirit and scope of the corresponding technical solutions of the embodiments of the present invention.

Finally, it should be noted that the present invention is not limited to the above alternative embodiments, and that various other forms of products can be obtained by anyone in light of the present invention. The above detailed description should not be taken as limiting the scope of the invention, which is defined in the claims, and which the description is intended to be interpreted accordingly.

Claims (10)

1. A construction method of a prediction model of engineering project bid quotation is characterized by comprising the following steps:

extracting the characteristics of the data by using principal component analysis or hierarchical clustering,

constructing an initial multiple linear regression model,

screening the initial model by utilizing the R-squared value, the P value and the average absolute error MAE and establishing an optimal prediction model;

and performing hypothesis test on the optimal model.

2. The method as claimed in claim 1, wherein the feature extraction is performed by principal component analysis.

3. The method for constructing the engineering project bid quotation prediction model according to claim 2, characterized in that the specific method for feature extraction is as follows:

standardizing the original data, and converting the original data into data with an average value of 0 and a standard deviation of 1;

calculating a covariance matrix of the normalized data;

calculating an eigenvalue and an eigenvector of the covariance matrix;

sorting the eigenvalues in a descending order, and then sorting the corresponding eigenvectors in a descending order;

and (5) extracting the characteristic vector corresponding to the characteristic value with the maximum first 2 values, drawing a biplot graph, and finding out the characteristic variable which is relatively close to the target variable in the graph.

4. The method for constructing the engineering project bid quotation prediction model according to claim 1, characterized in that the feature extraction by hierarchical clustering specifically comprises:

preprocessing the data, generally converting the original data into data with a mean value of 0 and a standard deviation of 1;

measuring the distance of a pair of data or data between two categories in the processed data;

aggregating two groups of data or two types of data according to the difference of distance categories;

the number of classes can be obtained by passing a horizontal straight line through the dendrogram, and the characteristic variables of the same class can be classified into one class.

5. The method of claim 1, wherein the engineering project bid quotation prediction model is constructed,

in the initial model, an R-squared value, a P value and an average absolute error MAE are adopted to screen the initial model, and the characteristic variables with the P value larger than 0.05 and the models with relatively smaller R-squared values are removed.

6. The method of claim 1, wherein the engineering project bid quotation prediction model is constructed,

substituting the independent variables in the test data into the prediction model to obtain a predicted value;

calculating the average absolute error MAE of the predicted value and the corresponding true value;

and selecting a model with a smaller average absolute error MAE as an optimal model.

7. The method as claimed in claim 1, further comprising performing hypothesis verification on the optimal prediction model by using a linear correlation test, a normal distribution test and a homodialogue test.

8. The construction method of the engineering project bid quotation prediction model as claimed in claim 1, wherein the extracted feature variables comprise price limit inventment, bridge length lofBridge, bridge proportion bridge prop, lane Nofroad, average quotation Avgoffer;

the initial prediction model is: avgoffer ═ inventment + lofBridge + bridgeProp + Nofroad;

the optimal prediction model is: avgoffer ═ inventstment + inventstment: bridgeProp, where inventstment: bridgeProp is the interaction of inventstment and bridgeProp.

9. A road and bridge project bid quotation prediction model, characterized in that, it is constructed by the method of any claim 1-8, the prediction model is: the method comprises the following steps of Avgoffer, inflstment and bridge, wherein Avgoffer is an average quote, inflstment is a limit price, bridge is a bridge proportion, and inflstment, bridge and bridge are interaction of inflstment and bridge.

10. A road bridge project bid quotation prediction method is characterized by comprising the following steps:

acquiring price limit inventment and bridge proportion bridge Prop data;

substituting the price limit inventment, bridge proportion bridgeProp data into the predictive model of claim 9.

Images